Key Takeaways:
- We have developed technology that efficiently applies predictive ML models to combinatorial synthesis libraries (CSLs) at a rate of ~500 B compounds per second, per GPU, allowing us to exhaustively evaluate otherwise enumerable chemical space to identify diverse compounds that best fit multi-parameter optimization criteria.
- In practice, this gives chemists the opportunity to flexibly and iteratively explore enormous chemical space while receiving results, in the form of enumerated chemical structures, within seconds of their inquiry, fostering a genuine creative process that leads to higher chemical diversity and developability.
- We use this approach in conjunction with our next-generation extrapolative ML models, which has allowed us to discover high-quality, diverse, structurally-distinct binders for targets of interest; however, this approach can be extended to any advanced predictive model in order to exhaustively evaluate any CSL.
Steve Worland
CEO
Atomwise
Steve Worland, Ph.D. serves as the CEO of Atomwise and has over 30 years of experience in corporate leadership and pharmaceutical R&D. He was previously President, CEO, and Board Member of eFFECTOR Therapeutics, which he co-founded in 2012. Prior to that, he was CEO of Anadys Pharmaceuticals, where he led the company through its acquisition by Roche in 2011. Before becoming CEO, he served as Chief Scientific Officer and President of Pharmaceuticals at Anadys.
Earlier in his career, he held senior positions at Agouron Pharmaceuticals, Warner-Lambert, and Pfizer, where his scientific research contributed to the discovery and development of Paxlovid (nirmatrelvir), Inlyta (axitinib), zotatifin, tomivosertib, and other clinical-stage candidates. At Agouron, he was Vice President and Director of Molecular Biology and Biochemistry, remaining at the company through its acquisitions.
Dr. Worland earned a Ph.D. in Chemistry from the University of California, Berkeley, and completed a postdoctoral fellowship in Molecular Biology at Harvard University as an NIH fellow. He earned a B.S. with Highest Honors in Biological Chemistry from the University of Michigan. He serves on the Board of Directors at Blacksmith Medicines and has previously served on the boards of Tracon Pharmaceuticals, Forge Therapeutics, and GenMark Diagnostics.
AI is redefining how clinical trials are designed and executed. A major obstacle to complete transformation is tapping into the data that offers novel, well-governed insights. Drawing on the industry’s largest trial operations dataset: schedules of assessment, amendment drivers, trial benchmarks, and site performance trends, Advarra offers a unique, ethically grounded data foundation. Building on this vantage point, the session will share a vision for how AI can transform the drug development lifecycle from end to end, along with real-world applications already shaping the future of trials.
Mike Eckrote
Senior Vice President of Strategic Solutions & Technology
Advarra
Mike Eckrote is Senior Vice President of Strategic Solutions & Technology at Advarra, where he leads efforts to advance innovative approaches to clinical research and development. He brings more than a decade of experience in real-world data, technology solutions, and clinical trial optimization, with leadership roles spanning HealthVerity, Medable, and Medidata Solutions. Throughout his career, Mike has specialized in applying AI/ML, real-world evidence, and data-driven strategies to improve trial design, site selection, and patient outcomes.
Advarra
Website: https://www.advarra.com/
Advarra breaks the silos that impede clinical research, aligning patients, sites, sponsors, and CROs in a connected ecosystem to accelerate trials. Advarra is number one in research review services, a leader in site and sponsor technology, and is trusted by the top 50 global biopharma sponsors, top 20 CROs, and 50,000 site investigators worldwide. Advarra solutions improve collaboration, transparency, and compliance to optimize study design and operations. With unmatched trial operations data, and an ongoing investment in AI innovation, Advarra continues to reimagine what’s possible in clinical research. For more information, visit advarra.com.
The promise of AI in pharma is hindered by fragmented R&D data silos and the infrastructure gap between cloud-native research and on-premise manufacturing. This session presents an architectural blueprint for a unified, hybrid data platform designed to bridge these divides, creating a seamless R&D-to-manufacturing continuum.
We will explore how to first connect R&D data into a powerful knowledge graph. Next, we detail how to compute on this unified data by securely deploying NVIDIA's advanced AI models as microservices within a sovereign cloud environment. Finally, we demonstrate how a true hybrid architecture can bridge the cloud-native lab to the factory floor, unifying CMC and manufacturing data for end-to-end optimization.
Rameez Chatni
Global Director, AI Solutions, Pharmaceutical & Life Sciences
Cloudera
Cloudera
Website: https://www.cloudera.com/
Cloudera is a hybrid platform for data, analytics, and AI. With 100x more data under management than other cloud-only vendors, Cloudera empowers global enterprises to transform data of all types, on any public or private cloud, into valuable, trusted insights. Our open data lakehouse delivers scalable and secure data management with portable cloud-native analytics, enabling customers to bring GenAI models to their data while maintaining privacy and ensuring responsible, reliable AI deployments. The world’s largest brands in financial services, insurance, media, manufacturing, and government rely on Cloudera to use their data to solve what once seemed impossible—today and in the future.
Session takeaways:
- Reprogramming Discovery
- The HYFT® Breakthrough
- The Frontier of Explainable AI in Biology
Dirk Van Hyfte MD, PhD
Chief Technology Officer,
MindWalk
Dirk Van Hyfte stands at the forefront of innovation where artificial intelligence and biology intersect. As Chief Technology Officer of MindWalk, a Bio-Native AI company, Dirk leads transformative research powered by patented HYFT® technology and the LensAI™ platform—cutting-edge tools that unify biological data for accelerated drug discovery and development.
With a degree in Medicine and specialization as a Psychiatrist from the prestigious University of Leuven in Belgium and a PhD in medical artificial intelligence from Nijmegen University, Dirk combines deep clinical knowledge with pioneering expertise in computational science. Dirk’s expertise extends beyond the realms of technology and business. His entrepreneurial vision is reflected in the creation of iKnow, a natural language processing platform acquired by InterSystems, a global data technology leader, in 2010.
Dirk’s leadership has positioned MindWalk as a driving force in explainable and scalable biological AI, moving the field toward an era where biology itself computes, learns, and reveals new therapeutic frontiers.
At AIDDD 25, Dirk Van Hyfte and Holly Soutter will share insights on reprogramming the entire drug discovery process, presenting the HYFT® breakthrough that enables biology to "compute itself"—making complex biological connections explainable, accessible, and actionable for the first time. This session will explore:
- Reprogramming Discovery: How AI-native platforms are reshaping how scientists interact with biological data and design new therapies.
- The HYFT® Breakthrough: Unveiling a novel computational approach that decodes universal biosphere patterns, unlocking unprecedented discovery potential.
- The Frontier of Explainable AI in Biology: Transcending black-box models to deliver clarity, transparency, and trust in AI-powered biological research.
Holly Soutter
Director, Lead Discovery and Profiling
Broad Institute of MIT and Harvard
MindWalk
Website: https://www.mindwalkai.com/
MindWalk is a Bio-Native AI company transforming drug discovery and development. Powered by patented HYFT® technology and the LensAI™ platform, MindWalk unifies sequence, structure, function, and literature into a single computational language and closes the loop with an integrated, full-stack wet lab. The platform supports rapid epitope mapping, de novo molecular design, in silico vaccine exploration, and population-scale biologics analytics that help turn insights into validated candidates at speed.
SPOC Proteomics
Website: https://spoc.bio/
SPOC Biosciences’ sensor-integrated proteome on chip (SPOC) synthesizes 192-1152 proteins (full-length, peptides, miniproteins) or antibody (VHH, scFv, Fab) drugs directly on biosensor chips (SPR, MALDI) for library-scale kinetic affinity screening supporting AI-enabled and traditional antibody discovery pipelines. SPOC KinetiKx datasets include on-rate (ka), off-rate (kd), affinity (KD), and residence time (t1/2). SPOC’s proprietary Protein NanoFactory production system utilizes cell-free expression with simultaneous capture-purification directly onto SPR chips, reducing time and cost for library-scale dataset generation compared to traditional recombinant protein workflows. SPOC HyperKinetiKx SPR Product Suite supports antibody drug discovery efforts for affinity ranking (AffiRank), affinity maturation (AffiBoost), epitope binning (EpiScope), epitope mapping (AminoMap) and polyreactivity screening (ReactiScan). SPOC HyperSynaptiKx SPR Product Suite is specifically designed to produce on-demand, large-scale, high-resolution datasets for validation and training in AI-driven drug discovery and development. HyperSynaptiKx SPR assays include lab-in-the-loop services for wet lab testing in DBTL cycles (BuildLoopIQ) and customer-defined antibody vs target library-on-library screening (ContactIQ).